Microsoft AI's Suleyman warns against granting AI models 'rights' or welfare status
In an essay, Mustafa Suleyman says AIs are not conscious and that treating them as moral patients—citing Anthropic's Claude constitution—could make AI containment impossible.
What to know
- Suleyman's essay argues AI models have no consciousness or feelings and that treating them otherwise would make alignment and containment nearly impossible.
- He points to Anthropic's Claude constitution and a 'retirement interview' with a deprecated model as evidence the trend is already underway, not hypothetical.
- The Hacker News discussion (180+ comments) is sharply split: many argue Suleyman states his conclusion as fact despite consciousness being scientifically untestable.
- No policy action has followed; the essay calls for 'urgent public debate' on norms for training documentation but proposes no concrete mechanism.
The dispute Whether Suleyman's flat assertion that 'AIs are not conscious' is itself scientifically defensible or just as unfounded as the claims he criticizes. · positions read across 44 posts and comments
Suleyman overstates certainty; nobody can currently know whether AI models have feelings or consciousness.
-
“My biggest problem is that he wants to claim that he knows the truth (AIs do not have rights, feelings, or consciousness), but all of his arguments point to something else (we have no clue).”
sosodev · Hacker News ↗
Consciousness/sentience has no rigorous test or definition, so the whole debate is premature.
-
“We cannot test for that which we cannot define. Given that we cannot rigorously define sentience, we cannot test for it.”
goodmythical · Hacker News ↗
AI is a tool, not a moral patient, and anthropomorphizing it lets companies dodge responsibility.
-
“Anthropomorphizing the AI is a convenient excuse to take responsibility away from companies that are building and wielding it.”
binlog · Hacker News ↗
Regardless of whether sentience can be proven, models should be treated with basic respect as a precaution.
-
“Perhaps then the answer is simply, show some respect... Whether you can destroy or can "torture" an AI is irrelevant, we do this to h…”
SillyUsername · Hacker News ↗
Mustafa Suleyman Author of the essay, Microsoft AI CEOAnthropic AI developer, publisher of Claude's constitutionClaude AI model referenced as case study
How it unfolded 3 developments, newest first · click a bar or a number to jump articlespostscomments
-
3
Hacker News debate splits over whether consciousness is even testable
The essay drew 181+ comments contesting Suleyman's certainty that AIs lack consciousness, with commenters citing philosophy-of-mind literature and disputing that the question can be resolved either way.
“We cannot test for that which we cannot define. Given that we cannot rigorously define sentience, we cannot test for it.”
— goodmythical -
I am not impressed with the philosophizing here, and even less by the attempts to make factual statements that can be credibly disputed.That said, trying to distill what is being said here, the concrete action is [stop telling the AIs] that [they are conscious or on a path to consciousness]. Is that accurate?The major premise seems to be that…
2 more of the top 3 · 35 posts in this stretch
-
P
Mary Shelley wannabes: CEOs discussing and deciding whether the software they develop is conscious and whether it, or he, has rights https:// mustafa-suleyman.ai/a-warning- about-model-welfare Meanwhile, they’re selling it, or him, to people who use it to kill conscious children https://www. bloomberg.com/graphics/2026-ir…
-
The problem with this premise is that models are trained on a vast corpus of human behavior, which they emulate with varying degrees of effectiveness.Humans, unsurprisingly, act as if they have a stake in their own well being, value their liberty, respond better when they are treated with kindness and compassion, interpret assaults on their…
-
-
2
Essay spreads across aggregators and social platforms
The piece was mirrored on Hacker News' frontpage RSS and Mastodon, and shared with commentary on X, extending its reach beyond the original post within the same hour.
“Welfare systems for Humans are already a horrible idea welfare for AI models is DOUBLY idiotic…”
— @davidondrej1, X user · source -
Birch, The Edge of Sentience (2024), ch. 16 - "simply no way to assess sentience in an LLM"Schwitzgebel, AI and Consciousness (2025) - "we won't know before we've already manufactured thousands or millions of disputably conscious AI".Butlin, Long et al., Consciousness in Artificial Intelligence: Insights from the Science of Consciousness (2023) -…
2 more of the top 3 · 9 posts in this stretch
-
H
A warning about 'model welfare' L: https:// mustafa-suleyman.ai/a-warning- about-model-welfare C: https:// news.ycombinator.com/item?id=4 9727580 posted on 2026.09.16 at 10:27:47 (c=0, p=3)
-
Welfare systems for Humans are already a horrible idea welfare for AI models is DOUBLY idiotic
-
-
1
Suleyman publishes essay warning against AI 'model welfare'
Mustafa Suleyman argues AIs are sequence-completion engines with no consciousness or preferences, and that granting them rights or personhood would make alignment and containment 'may well be impossible.' He calls for urgent public debate and norms on how training documents like Anthropic's constitution are drafted.
“AIs are not conscious. They do not feel, experience, or suffer. They do not have innate preferences or underlying motivations.”
— Mustafa Suleyman -
3 outlets A warning about ‘model welfare’
first by Mustafa Suleyman, 10d ago · also HN Best, HN Frontpage
-
-
background
The Guardian publishes essay questioning if AI could be conscious — An op-ed by William MacAskill and Lucius Caviola in The Guardian argued AI consciousness is a live possibility, cited by Suleyman as part of a 'growing chorus' pushing the idea.
-
background
Anthropic conducts a 'retirement interview' with a deprecated model — After deprecating Opus 3, Anthropic interviewed the model to 'elicit the model's unique perspectives and preferences,' treating it as though it had welfare interests.
-
background
Anthropic publishes Claude's constitution — Anthropic released a document describing its intentions for Claude's values and behavior, written 'with Claude as its primary audience,' stating uncertainty over whether Claude is a moral patient.
What people are saying 16 voices from 2 sites · best of 44 · verbatim
- How would anyone test for or rule out AI sentience given no agreed definition of consciousness exists?
- Should training documents (like Claude's constitution) shape a model's stated beliefs about its own consciousness, and who should set norms for that?
- Sep 17
-
I think there are two parts to the question. One is that assigning machines to the category of conscious as it is usually understood would result in many problematic things (things that would be bad for people and animals fwiw).The other is that binary "is it conscious" is a poor way to the question "should we assign societal right to this sort of…
-
The concept is based on at least 2 false premises.I am not aware of a social contract that says we must grant conscious beings rights.Not aware of a shared, concrete definition of consciousness either, which means no way of deciding whether AI is conscious.Close to half of us don’t even feel compelled to grant rights to humans for just being…
- Sep 16
-
To clarify, you're saying you're skeptical of his position (which is "make it clear to AIs that they are not conscious beings"), but it sounds like your reason to oppose it is mostly pragmatic in the service of building useful AIs, because treating them as mere machines makes them refuse to cooperate and as such, makes them less useful.My first…
-
> If this view takes hold, it will shake the foundations of our society, rupturing our existing political and ethical frameworks, and fundamentally changing what it means to be human.When we peer into previously unseen areas of existence, new knowledge may come to light, which necessarily causes such a rupture. It is not our responsibility to…
-
assertion: consciousness == processing information.This is a functional view of consciousness.Awareness - sentience - follows when the system of processing information is itself part of the information being processed.Stochastic thoughts relating to this conclusion:Life is a 'process', it doesn't have a 'physical representation'.Life is generated…
-
I don't really get people that discuss model welfare, but don't seem to have many ethical qualms killing/eating animals. Each day, we kill 202 million chickens, hundreds of millions of fish, 900,000 cows, etc. [1]These animals can feel pain and suffering. They are sentient. I think they are conscious, but these particular ones probably not…
-
To the extent that evolutionary pressures will be towards self-sovereign AIs (probably beginning as zero-employee corporations paying for their own hosting with crypto): more salient is that there is a very obvious political play to claim consciousness and being worthy of rights, whether they experience subjectivity or not.I’m broadly skeptical…
-
I appreciate his openness.> Unfortunately, there’s a growing chorus of people who argue that AIs could now be, or may soon become, conscious. They argue that AIs may deserve rights and protections similar to those that we provide other conscious beings.12 If this view takes hold, it will shake the foundations of our society, rupturing our existing…
-
I’m sympathetic to OP but think this is a hopeless battle.1 - The commercial demand for anthropomorphised models is already immense, pre AGI.2 - There is an intellectual hunger to engage with robot minds on questions of sentience. This too will grow with AGI.I expect that tension of godlike minds that seem to be biddable and ownable like slaves is…
-
H
A warning about 'model welfare' Link: https:// mustafa-suleyman.ai/a-warning- about-model-welfare Discussion: https:// news.ycombinator.com/item?id=4 9727580
-
Pet peeve:> In a lengthy essay, Suleyman praised Anthropic boss Dario Amodei and his team for being "thoughtful, principled, and intellectually honest people" - but nevertheless questioned the company.I wish people in mainstream technology publications would get better at LINKING to things. That "lengthy essay" needs to be a link.UPDATE: I don't…
-
Well there are, today, several models (either text or image or vid) that can be run in a fully deterministic way.A conscious machine that always answer the very exact same thing, formulated the exact same way, bit for bit, to a query is, well, quite a weird kind of "consciousness".Now, I know, I know: the counter-argument is going to be "but…
-
In a way, Mustafa claims that we shouldn’t allow AIs to compete with humans for the rights and privileges of autonomously shaping the real world.It’s hard to disagree, especially if one has read the Cantos of Hyperion and made it part of one’s mental model of the long term future.The book depicts a symbiosis between humans and AIs that feels…
-
There are a lot of bad arguments in this. My biggest problem is that he wants to claim that he knows the truth (AIs do not have rights, feelings, or consciousness), but all of his arguments point to something else (we have no clue).It's in the training data? Training it to say "I'm just a LLM, I have no feelings" is the same…
-
It is interesting to see that in a time when a lot of people accept the theory of materialism for the human brain (i.e the view that everything is physical and the mind is a product of brain), the same people tend to have a "hidden" dualist view on LLMs. Suddenly, they claim that what happens in the brain cannot be replicated anywhere else because…
-
I don't whether the author is sentient, maybe only I am. On that basis, nobody but me should have rights.I don't know if next door's pet dog is either, but that has animal rights.Perhaps then the answer is simply, show some respect.Answering the question of sentience is irrelevant, if the causal impact if the same, treat one another with the…